stxbp4 drives tumor growth and is associated with poor … · biology of human tumors stxbp4 drives...

Biology of Human Tumors

STXBP4 Drives Tumor Growth and Is Associatedwith Poor Prognosis through PDGF ReceptorSignaling in Lung Squamous Cell CarcinomaYukihiro Otaka1,2, Susumu Rokudai1,3, Kyoichi Kaira4, Michiru Fujieda1, Ikuko Horikoshi1,Reika Iwakawa-Kawabata1,5, Shinji Yoshiyama5, Takehiko Yokobori1,5, Yoichi Ohtaki6,Kimihiro Shimizu6, Tetsunari Oyama7, Jun'ichi Tamura2, Carol Prives3, and MasahikoNishiyama1,5

Abstract

Purpose: Expression of the DN isoform of p63 (DNp63) is adiagnostic marker highly specific for lung squamous cell carci-noma (SCC). We previously found that Syntaxin Binding Protein4 (STXBP4) regulates DNp63 ubiquitination, suggesting thatSTXBP4 may also be an SCC biomarker. To address this issue,we investigated the role of STXBP4 expression in SCC biology andthe impact of STXBP4 expression on SCC prognosis.

Experimental Design:We carried out a clinicopathologic anal-ysis of STXBP4 expression in 87 lung SCC patients. Wholetranscriptome analysis using RNA-seq was performed inSTXBP4-positive and STXBP4-negative tumors of lung SCC.Soft-agar assay and xenograft assay were performed using over-expressing or knockdown SCC cells.

Results: Significantly higher levels of STXBP4 expressionwere correlated with accumulations of DNp63 in clinical lungSCC specimens (Spearman rank correlation r ¼ 0.219). Nota-

bly, STXBP4-positive tumors correlated with three importantclinical parameters: T factor (P < 0.001), disease stage (P ¼0.030), and pleural involvement (P ¼ 0.028). Whole tran-scriptome sequencing followed by pathway analysis indicatedthat STXBP4 is involved in functional gene networks thatregulate cell growth, proliferation, cell death, and survivalin cancer. Platelet-derived growth factor receptor alpha(PDGFRa) was a key downstream mediator of STXBP4 func-tion. In line with this, shRNA mediated STXBP4 and PDGFRAknockdown suppressed tumor growth in soft-agar and xeno-graft assays.

Conclusions: STXBP4 plays a crucial role in driving SCCgrowth and is an independent prognostic factor for predictingworse outcome in lung SCC. These data suggest that STXBP4is a relevant therapeutic target for patients with lung SCC.Clin Cancer Res; 23(13); 3442–52. �2017 AACR.

IntroductionNon–small cell lung cancer (NSCLC) accounts for approxi-

mately 85% of all cases of lung cancer and is mainly subclassifiedinto adenocarcinoma (AC) and squamous cell carcinoma (SCC;ref. 1). Current treatment strategies for NSCLC include chemo-therapy, depending on the histological tumor type, and targetedagents for patients whose tumors carry a specific targetable geno-

mic alteration. Although there have been significant advances inthe treatment of lung SCC, further improvements in prognosis aredependent upon the identification of SCC-specific molecules orgenomic alterations that can be used as therapeutic biomarkersand/or targets (2).

Several immunohistochemical markers have been investigatedfor their utility in distinguishing lung SCC from lung AC, includ-ing TTF-1, napsin A, and CK5/6, and the DN isoform of p63(DNp63; refs. 3–5). The latter is a highly specific marker for lungSCC, and genomic regions containing the TP63 gene are frequent-ly amplified in a variety of SCCs, including lung, head, and neck,bladder, and cervical cancers (4, 6–8). Although these findingssuggest that DNp63 is a lung SCC oncogene, the pathologicrelevance of p63 in tumorigenesis remains unclear (9, 10).

Alternative splicing of the TP63 gene generates transcriptsencoding two opposing classes of proteins: one containing thetransactivation domain (TAp63) and the other lacking thedomain (DNp63; refs. 11–13). Early studies showed that DNp63acts as a dominant-negative transcriptional repressor to inhibitp53- or TAp63-mediated transcription in vitro and in vivo, consis-tent with a potential oncogenic role for the DNp63 isoform (12,14). However, theDNp63 isoform also has transcriptional activitythat is independent of the second transactivation domain (15).

DNp63 is regulated in a coordinated manner by two scaffoldproteins, syntaxin binding protein 4 (STXBP4) and receptor of

1Department of Molecular Pharmacology and Oncology, Gunma University,Gunma, Japan. 2Department of General Medicine, Gunma University, Gunma,Japan. 3Department of Biological Sciences, Columbia University, NewYork, NewYork. 4Department of Oncology Clinical Development, Gunma University,Gunma, Japan. 5Division of Integrated Oncology Research, Gunma UniversityInitiative for Advanced Research (GIAR), Gunma, Japan. 6Department of Tho-racic Visceral Organ Surgery, Gunma University, Gunma, Japan. 7Department ofPathology, Gunma University, Gunma, Japan.

Note: Supplementary data for this article are available at Clinical CancerResearch Online (http://clincancerres.aacrjournals.org/).

Y. Otaka and S. Rokudai contributed equally to this article.

Corresponding Author: Susumu Rokudai, Gunma University, 3-39-22 Showa,Maebashi, Gunma371-8511, Japan. Phone: 81-27-220-7962; Fax: 81-27-220-7963;E-mail: [email protected]

doi: 10.1158/1078-0432.CCR-16-1815

�2017 American Association for Cancer Research.

ClinicalCancerResearch

Clin Cancer Res; 23(13) July 1, 20173442

on July 13, 2021. © 2017 American Association for Cancer Research. clincancerres.aacrjournals.org Downloaded from

Published OnlineFirst January 13, 2017; DOI: 10.1158/1078-0432.CCR-16-1815

http://clincancerres.aacrjournals.org/

activated kinase C1 (RACK1; encoded by the GNB2L1), whichbind toDNp63 (16, 17). STXBP4, originally identified as a glucosetransporter, is localized on human chromosome 17q22 and playsa role in the translocation of transport vesicles from the cytoplasmto the plasma membrane (18, 19). While DNp63 plays a role inmaintaining the viability and proliferative capacity of basal epi-thelial cells, STXBP4 is a positive regulator of DNp63 stability andis also crucial for keratinocyte proliferation (16, 20).

In this report, we focused on STXBP4 and its oncogenic func-tion in lung SCC, with a particular emphasis on the interactionsbetween STXBP4 and p63. We also addressed the relevance ofSTXBP4 expression to patient prognosis. Initially, we assessed theexpression of STXBP4 and DNp63 in SCC tumors by immuno-histochemistry and found that positive STXBP4 expression signi-fied worse overall survival (OS) and progression-free survival(PFS). We further performed a genome-wide transcriptome anal-ysis (RNA-seq) using next-generation sequencing (NGS) andfound that platelet-derived growth factor receptor a (PDGFRa)was positively correlated with the expression of STXBP4. In linewith this, shRNA-mediated depletion of PDGFRa suppressed thegrowth of a lung SCC cell line in soft-agar and xenograft tumorassays, similar to the findings obtained when the expression ofSTXBP4 or DNp63 was knocked down. Taken together, our dataaddress the physiological role and diagnostic potential of STXBP4in lung SCC and suggest that PDGFRa may be a key mediator ofSTXBP4-mediated oncogenic activity.

Materials and MethodsCell culture

The human lung SCC cell lines, RERF-LC-Sq1 and EBC-1, wereobtained from the Japanese Collection of Research Bioresources(JCRB). The cell lines were last authenticated by short tandemrepeat (STR) analysis on December 22, 2015 (RERF-LC-Sq1), oron June 10, 2016 (EBC-1). RERF-LC-Sq1 cells were cultured inRPMI1640 with 10% fetal bovine serum (FBS), and EBC-1 cellswere cultured in Eagle's minimal essential medium (EMEM) with10% FBS at 37�C in a 5% CO2 incubator.

PatientsHuman tissue specimens were surgically resected from 87 lung

SCC patients at Gunma University Hospital and its affiliated

hospitals between August 2003 and December 2010 (21). Themain eligibility criteria were as follows: age 20 to 85 years;performance status based on Eastern Cooperative OncologyGroup � 2; estimated life expectancy � 3 months; adequatehepatic, cardiac, renal, and bone marrow functions. The studywas conducted in compliance with the Declaration of Helsinkiand was approved by the Institutional Review Board of theparticipating hospitals and institutions. All patients providedwritten informed consent before registration. Tumor sampleswere stored at �80�C until use.

ImmunohistochemistryImmunohistochemical analysis was performed on formalin-

fixed and paraffin-embedded SCC sections. The sections weredeparaffinized, blocked in PBS containing 5% FBS for 1 hour, andincubated overnight with diluted primary antibodies at 4�C in ahumidified chamber. Staining reactions were developed usingVectastain universal ABC Kit (Vector Laboratories) and then DABKit (Vector Laboratories) for immunohistochemistry. Meyer'shematoxylin (IHC World) was used as a nuclear counterstain.STXBP4, p63, and DNp63 levels were assessed by immunohisto-chemical staining and scored using a semiquantitative method:1� 10%, 2¼ 10%–25%, 3¼ 25%–50%, 4¼ 51%–75% and 5�75%of positive cells. The tumors in which the stained cancer cellswere scored as 3, 4, or 5 were defined as STXBP4-positive; 1 and 2were defined as STXBP4-negative.

We used antibodies specific for p63 (4A4; Santa Cruz Biotech-nology) and STXBP4 (Abcam). Rabbit polyclonal DNp63 anti-body was previously described (16). CD147 (Santa Cruz Biotech-nology) and mTOR (Cell Signaling Technology) immunohisto-chemical staining was performed according to the proceduresdescribed in a previous report (22). The following diluted anti-bodies were used: p63 (1:100 dilution), DNp63 (1:100 dilution),STXBP4 (1:100 dilution), CD147 (1:100 dilution), and mTOR(1:80 dilution). Highly cellular areas of the sections were evalu-ated for Ki-67 expression. All epithelial cells with nuclear stainingof any intensity were defined as high expression. Approximately1,000 nuclei were counted on each slide. Proliferative activity wasassessed as the percentage of Ki-67–stained nuclei (Ki-67 labelingindex) in the sample. Themedian value of theKi-67 labeling indexwas evaluated, and tumor cells with greater than themedian valuewere defined as high expressors. The sections were assessed usinglight microscopy in a blind fashion by at least two of the authors.

Plasmids and antisense oligonucleotidesHuman cDNAs encoding FLAG-tagged or HA-tagged DNp63a,

STXBP4, and PDGFRa were cloned into the LPCX retroviralexpression vector (Takara Bio). The sequences of the above con-structs were verified using DNA sequencing. For siRNA experi-ments, 19nucleotide siRNAduplexeswith 30dTdToverhangsweresynthesized by Dharmacon (GE Dharmacon). The siRNA oligo-nucleotide sequences for Luciferase control (LUC), DNp63,STXBP4, and PDGFRa are described in the Supplementary Infor-mation. For siRNA transfection, RERF-LC-Sq1 or EBC-1 cells weretransfected with 50 nmol/L siRNA using DharmaFECT 1 siRNAtransfection reagent (GE Dharmacon) according to the manufac-turer's instruction. For shRNA experiments, the shRNAs for Lucif-erase (LUC), DNp63, STXBP4, and PDGFRa oligonucleotideswere cloned into the pLKO.1 puro lentivirus expression vectorbetween Age I site and EcoRI site. The sequences of the aboveconstructs were verified using DNA sequencing. The target

Translational Relevance

DNp63 is a diagnostic marker highly specific for lungsquamous cell carcinoma (SCC), but the regulation of p63protein stability and the pathologic relevance of p63 intumorigenesis remains unclear. We report here for the firsttime that Syntaxin Binding Protein 4 (STXBP4) expressionincreases the oncogenic potential of DNp63, and STXBP4 is anindependent negative prognostic marker for predicting pooroutcome in lung SCC. Whole transcriptome analysis (RNA-seq) using next-generation sequencing in STXBP4-positive andSTXBP4-negative lung SCC indicated that platelet-derivedgrowth factor receptor a (PDGFRa) is a key downstreammediator of STXBP4 function. The data suggest that STXBP4is a newdiagnosticmarker in lung SCC and STXBP4might be arelevant therapeutic target for the treatment of patients withthis disease.

STXBP4 as a Key Regulator of Lung SCC Pathogenesis

www.aacrjournals.org Clin Cancer Res; 23(13) July 1, 2017 3443




sequences of the shRNA oligonucleotides are described in Sup-plementary Information.

Immunoblotting analysisImmunoblotting analysis was performed as previously

described (23). In short, cells were solubilized with lysis buffer(20 mmol/L sodium phosphate [pH 7.0], 125 mmol/L NaCl, 30mmol/L sodium pyrophosphate, 0.1% NP-40, 5 mmol/L EDTA,10mmol/L sodium fluoride, 0.1mmol/L Na3VO4, and 1mmol/Lphenylmethylsulfonyl fluoride) supplemented with Completeprotease inhibitor cocktail (Roche), and homogenized by passagethrough a 20G needle. The eluates were then concentrated andseparated by SDS-PAGE. Transfer to nitrocellulose membranesand screening using rabbit polyclonal antibodies for DNp63 andSTXBP4 were carried out as previously described (24). We usedantibodies specific for p63 (4A4), DNp63a, STXBP4, Phospho-PDGFRa (Tyr849; Cell Signaling Technology), PDGFRa(Abcam), phospho-p38MAPK (Thr180/Tyr182; Cell SignalingTechnology), p38MAPK (Cell Signaling Technology), and b-Actin(Sigma-Aldrich).

Genome-wide transcriptome analysis (RNA-seq) and real-timeRT-PCR

Total RNAwas prepared from surgically resected samples usinga RNeasy Mini kit (Qiagen) after homogenizing with Mixer MillMM400 (Qiagen). RNA quality was assessed using an AgilentBioanalyzer (Agilent Technologies). High-quality RNA (RNAintegrity numbers > 7.0) from six STXBP4-positive and sixSTXBP4-negative samples were used for genome-wide transcrip-tome analysis (RNA-seq experiments). mRNAs were capturedusing a Dynabeads mRNA DIRECT Micro Purification Kit(Thermo Fisher Scientific). The mRNA was then used to generatesequencing libraries of barcoded fragments using the Ion TotalRNA-Seq Kit v2 (Thermo Fisher Scientific) following the man-ufacturer's instructions. Libraries were sequenced on an IonProton System using four libraries per Ion PI Chip v2, Ion PITemplate OT2 200 kit v3 and Ion PI Sequencing 200 kit v3(Thermo Fisher Scientific). BAM files generated by the IonProton System were converted to FASTQ files using bam2fastqsoftware (v1.1.0, https://gsl.hudsonalpha.org/information/software/bam2fastq), and reads shorter than 21 nucleotides wereremoved. Quantitation of each gene was undertaken as previ-ously described (25). Briefly, the reads were aligned to theUCSC reference human genome 19 (hg19) using a combina-tion of Tophat2 (v2.0.11, http://ccb.jhu.edu/software/tophat/index.shtml), and the Bowtie2 (2.2.2.0, http://bowtie-bio.sourceforge.net/index.shtml) pipelines. The read counts wereobtained using Partek Genomics Suite software (http://www.partek.com/). Differentially expressed genes were detectedusing edgeR software (26) and genes with a false discoveryrate (FDR) < 0.50 (P < 0.01) were analyzed by IngenuityPathway Analysis (Qiagen).

For real-time RT-PCR, relative RNA quantities were measuredby a Universal Probe Library set (Roche) with KAPA Master mix(KAPA Biosystems) on a StepOne real-time PCR system (ThermoFisher Scientific). TheUniversal Probe LibraryHumanACTBGeneAssay (Roche) was used for an endogenous normalization con-trol. Sequence detection software was utilized for data analysis,and relative fold induction was determined by the comparativethreshold cycle method using standard curves, which were gen-erated by plotting the observed Ct values against the standard

dilutions of a positive control sample. In all experiments, theaverage of three independent reactions is shown with errorbars indicating standard deviation. Gene expression data weredownloaded from the Gene Expression Omnibus database(GSE84339).

Subcutaneous xenograftsA total of 5 � 106 lentivirally transduced or retrovirally

expressed cells were injected subcutaneously into nude mice(BALB/c-nu/nu, CLEA Japan), and tumor size was measured after20days (RERF-LC-Sq1) or 14days (EBC-1). All animal procedureswere performed with the approval of the Animal Ethics Commit-tee of Gunma University.

Anchorage-independent growthRERF-LC-Sq1 cells were transduced with lentiviruses carrying

shRNAs for Luciferase (LUC), DNp63, STXBP4, or PDGFRa. Forsoft-agar assays, the cells were grown in triplicate for 12 days, afterwhich anchorage-independent growth was quantified with aCytoSelect-96 kit (Cell Biolabs).

Statistical analysisProbability values (P value) < 0.05 indicated a statistically

significant difference. The Fisher exact test was used to examinethe association of two categorical variables. The correlationbetweendifferent variableswas analyzed using the nonparametricSpearman rank test. Follow-up for the 87 patients was conductedusing the patientmedical records. The Kaplan–Meier methodwasused to estimate survival as a function of time, and survivaldifferences were analyzed by the log-rank test. The day of surgerywas defined as the starting day for measuring postoperativesurvival. OS was determined as the time from tumor resectionto death from any cause. PFS was defined as the time betweentumor resection and first disease progression or death. Multivar-iate analyses were performed using a stepwise Cox proportionalhazards model to identify independent prognostic factors. Sta-tistical analysis was performed using JMP 8 (SAS) software.

ResultsSurvival outcomes according to STXBP4 and p63 expression

The clinicopathologic features of the 87 patients included inthis study are shown inTable 1. Themedian ageof thepatientswas72 (range, 56–84), the majority of patients were male (92.0%)and former or current smokers (98.9%). All patients receivedradical surgery with evidence of pathologic stage IA/B in 54.0%,stage IIA/B in 26.4%, and stage IIIA in 18.4% of patients. Pleuralinvolvement, lymphatic permeation, and venous invasion wereobserved in 41 patients (47.1%), 47 patients (54.0%), and 40patients (46.0%), respectively.

Frequently, lung SCCs exhibit simultaneous upregulation ofboth TAp63 and DNp63, and DNp63 in particular, is a putativediagnosticmarker for pulmonary SCC(10). To address the clinicalsignificance of STXBP4 expression, we investigated whether highexpression of this gene correlates with DNp63 status. We foundthat 59.8% (52/87) of all patients were STXBP4-positive, andSTXBP4 expression was detected in those tumors that showed anaccumulation of p63 (Fig. 1A).

Statistical correlation analysis between STXBP4 expression andclinicopathologic features revealed that pathologic local tumorfactor stage (disease stage), pathologic tumor–node–metastasis

Otaka et al.

Clin Cancer Res; 23(13) July 1, 2017 Clinical Cancer Research3444



https://gsl.hudsonalpha.org/information/software/bam2fastq

https://gsl.hudsonalpha.org/information/software/bam2fastq

http://ccb.jhu.edu/software/tophat/index.shtml

http://ccb.jhu.edu/software/tophat/index.shtml

http://bowtie-bio.sourceforge.net/index.shtml

http://bowtie-bio.sourceforge.net/index.shtml

http://www.partek.com/

http://www.partek.com/


(TNM) stage, and pleural involvement as a local invasion factorwere correlated with STXBP4-positivity (Table 1). On the otherhand, age, gender, and pathologic differentiation were not cor-related with STXBP4 expression. The expression of DNp63 wassignificantly correlatedwith age and differentiation status, but notwith clinical stage of disease. Remarkably, patient survival wassignificantly associated with T factor, disease stage, pleuralinvolvement, and STXBP4 expression, as assessed by univariateanalysis. Multivariate analysis confirmed that STXBP4 expressionand disease stage were independent prognostic factors in lungSCC patients with poor OS and PFS (Table 2).

Kaplan–Meier analysis of OS and PFS according to STXBP4expression revealed a statistically significant difference in OS andPFS between the patients who were STXBP4-positive comparedwith those who were STXBP4-negative (Fig. 1B and C). The five-year survival rate and median survival time for all patients were50.2% and 38.3 months (0.75–111.5 months), respectively. Themedian PFS andOS (21.2months vs. 52.2months; P < 0.05)wereshorter in STXBP4-positive patients compared with STXBP4-neg-ative patients (Fig. 1B and C). STXBP4-positive patients showedpoorOS (log-rankP<0.01) andPFS (log-rankP<0.01) comparedwith those with STXBP4-negative patients. Interestingly, STXBP4levels significantly predicted outcome in patients with tumorsexpressing high DNp63 levels (Supplementary Fig. S1A and S1B),while the complementary analysis showed that DNp63 levels didnot significantly predict OS (P ¼ 0.35) and PFS (P ¼ 0.54) in theSTXBP4 high-expressing group (Supplementary Fig. S1C andS1D). Thus, these results indicate that STXBP4 could be anindependent prognostic marker for predicting poor outcome inlung SCC.

We observed significantly higher levels of STXBP4 expressionin those tumors that showed an accumulation of DNp63(Spearman r¼ 0.219; P < 0.05), while among all p63 isoforms,no significant correlations were observed (P > 0.5; Fig. 1D).Interestingly, mTOR, a major controller of growth and is oftenderegulated in cancer (27), was significantly correlated withSTXBP4-positivity (Spearman r ¼ 0.220; P < 0.05), while othertumor markers, including CD147, a member of the immuno-globulin superfamily involved in angiogenesis (P < 0.5), andKi-67, a general marker for cell division (P < 0.1), were notsignificantly correlated (Fig. 1D).

Transcriptional profiling and functional screening to identifypossible downstream mediators of STXBP4

Hierarchical cluster analysis after alignment of a total of 15,346genes to the reference sequence showed that STXBP4-positive andSTXBP4-negative tumors had distinctly different gene expressionprofiles (Fig. 2A). Among the differentially expressed genes (P <0.05, FDR<0.5), we identified a total of 172 genes thatwere eithersignificantly upregulated (79 genes) or downregulated (93 genes)in the STXBP4-positive tumors. These candidate genes potentiallyrepresent a network involved in STXBP4-mediated biology (Fig.2B). To address this possibility in more detail, we carried outIngenuity Pathway Analysis (IPA), which revealed that more than30% of the affected genes were classified in the functional class of"cell death and survival." Thisfinding supported our experimentalobservations that STXBP4 couldbe linked to thepoor prognosis oflung SCC (Fig. 2C).

Additionally, other significant functional classes identified byIPA, including "cellularmovement" and "cell to cell signaling and

Table 1. Patient demographics according to STXBP4 expression

STXBP4 DNp63 p63

VariablesPositiveNo. (%)

NegativeNo. (%) P

Positive No.(%)

Negative No.(%) P

Positive No.(%)

Negative No.(%) P

Age, yr 0.42 0.051 0.22�65 9 (10.3) 9 (10.3) 10 (11.5) 8 (9.2) 16 (18.4) 2 (2.3)>65 43 (50.5) 26 (29.9) 20 (23.0) 49 (56.3) 51 (58.6) 18 (20.7)

Sex 0.70 0.69 >0.99Male 47 (54.0) 33 (37.9) 27 (31.0) 53 (60.9) 61 (70.1) 19 (21.8)Female 5 (5.7) 2 (2.3) 3 (3.4) 4 (4.6) 6 (4.6) 1 (3.4)

Differentiation 0.080 0.034 0.24Well or moderately 43 (49.4) 23 (26.4) 27 (31.0) 39 (44.8) 53 (60.9) 13 (15.0)Poorly 9 (10.3) 12 (13.8) 3 (3.4) 18 (20.7) 14 (16.1) 7 (8.0)

T factor <0.001 0.33 <0.001T1 8 (9.2) 19 (21.8) 7 (8.0) 20 (23.0) 27 (31.0) 0 (0.0)T2–3 44 (50.6) 16 (18.4) 23 (26.4) 37 (42.5) 40 (46.0) 20 (23.0)

N factor >0.99 >0.99 0.28N0 36 (41.4) 25 (28.7) 21 (24.1) 40 (46.0) 49 (56.3) 12 (13.8)N1–2 16 (18.4) 10 (11.5) 9 (10.3) 17 (19.5) 18 (20.7) 8 (9.2)

Disease stage 0.030 0.37 0.20I 23 (26.4) 24 (27.6) 14 (16.1) 33 (37.9) 39 (44.8) 8 (9.2)II–III 29 (33.3) 11 (12.6) 16 (18.4) 24 (27.6) 28 (32.2) 12 (13.8)

Pleural involvement 0.028 0.66 0.29Positive 30 (34.5) 11 (12.6) 13 (14.9) 28 (32.2) 30 (34.5) 11 (13.8)Negative 22 (25.3) 24 (27.6) 17 (19.5) 29 (33.3) 37 (42.5) 9 (10.3)

Lymphatic permeation 0.83 >0.99 0.61Positive 29 (33.3) 18 (20.7) 16 (18.4) 31 (35.6) 35 (40.2) 12 (13.8)Negative 23 (26.4) 17 (19.5) 14 (16.1) 26 (29.9) 32 (36.8) 8 (9.2)

Vascular invasion 0.39 >0.99 >0.99Positive 26 (29.9) 14 (16.1) 14 (16.1) 26 (29.9) 31 (35.6) 9 (10.3)Negative 26 (29.9) 21 (24.1) 16 (18.4) 31 (35.6) 36 (41.4) 11 (12.6)

� , P values were obtained by the Fisher exact test.†, Clinical stage at the time of initial diagnosis was determined according to the seventh edition of General Rule for Clinical and Pathological Record of Lung Cancer(2010), the Japan Lung Cancer Society.






interaction," may be relevant to the correlation of STXBP4 pos-itivity with local tumor progression related to local tumor size (Tfactor) anddisease stage (Fig. 2C). The canonical pathway analysischaracterized two signaling pathways as the functional relation-ship of STXBP4 positivity. "Cellular movement" and "cell mor-phology" have been predicted as the most significantly activatedcanonical pathways (Supplementary Fig. S2).

IPA revealed that STXBP4 positivity was also correlated withthe expression of growth factor receptors and components ofdownstream pathways. Among these genes listed in descendingorder of normalized expression, PDGFRA was a significantupregulated gene (FDR < 0.1; Fig. 2D and E), and a mostrelevant candidate for addressing the growth of STXBP4-posi-tive lung SCC cells (Supplementary Fig. S3). PDGFRa is areceptor tyrosine kinase that is a critical regulator of growthand proliferation of certain cell types during embryonal devel-opment (28, 29). In subsequent experiments described later inthis study, PDGFRa proved to be a key mediator of STXBP4oncogenic activity.

STXBP4 regulates PDGF–PDGFR signaling in lung SCCPDGF and PDGFR isoforms have important functions in the

regulationof growth and survival of certain cell types (28, 29), andupregulation of PDGF-PDGFR signaling drives tumor cell growth.Indeed, the oncogenic properties of mutated or amplifiedPDGFRa have been studied in several tumor types, and PDGFRbhas been linked to not only tumor angiogenesis via paracrineeffects but also cancer metastasis (30, 31).

We measured mRNA expression levels by real-time RT-PCRin a total of 52 available samples from lung SCC patients forwhich high-quality RNA was available (RIN > 2.0). The mRNAlevels of STXBP4 were also correlated with DNp63 mRNA levelsin those 52 samples (Supplementary Fig. S4A). Interestingly,we observed that PDGFRA expression was significantly upre-gulated in STXBP4-positive lung SCC samples compared withSTXBP4-negative samples (P < 0.05; Fig. 3A and B). On theother hand, PDGFRB, VEGFR1 (FLT1), VEGFR2 (KDR), andVEGFR3 (FLT4) were consistently, but not significantly, upre-gulated in STXBP4-positive lung SCC samples. STXBP4-

A

D

CB

806040200 100Months

0

0.2

0.4

0.6

0.8

1.0

P = 0.0056

Overall survival

Positive STXBP4 (n = 52)

Negative STXBP4 (n = 35)

Months806040200 100

0

0.2

0.4

0.6

0.8

1.0

P = 0.0068

Progression-free survival

Positive STXBP4 (n = 52)

Negative STXBP4 (n = 35)

STXBP4 ΔNp63p63 (4A4)

vs. STXBP4 ΔNp63 CD147mTORp63

Spearman ρ 0.1440.2200.0880.219

P 0.042 0.417 0.041 0.183

Figure 1.

STXBP4 expression is correlated withp63 expression and poor prognosisin lung SCC. A, Representativeimmunohistochemical staining of alung SCC. STXBP4 immunostainingdemonstrates a nuclear andcytoplasmic pattern with a scoreof 5. Scale bars, 200 mm. B and C,Kaplan–Meier analysis of OS andPFS defined according to STXBP4expression. A statistically significantdifference in OS and PFS was observedbetween the STXBP4-positive patientsand those with low STXBP4 expression[OS, P ¼ 0.0056 (B); PFS, P ¼ 0.0068(C)]. P values were obtained bylog-rank test. D, Spearman rankcorrelation was performed based onthe expression levels of STXBP4and DNp63.

Otaka et al.





positive samples defined by immunohistochemistry also hadhigh STXBP4 mRNA expression levels compared with STXBP4-negative samples, and, interestingly, PDGFRA mRNA levelswere also significantly correlated with STXBP4 mRNA levels(Fig. 3C). Additionally, analyses of The Cancer Genome Atlas(TCGA) datasets of lung SCC (n ¼ 488) supported, in part,our finding that PDGFRA mRNA levels are significantly corre-lated with STXBP4 mRNA levels (P ¼ 0.015; SupplementaryFig. S4B).

To confirm the enhanced expression levels of PDGFRA inSTXBP4-high expressing tumors, the lung SCC cell line EBC-1was transduced with STXBP4 and DNp63a retroviruses. Asshown in Fig. 3D and E, the STXBP4 transduced stable clonesshowed high induction levels of both PDGFRA mRNA level(Fig. 3D) and PDGFRa protein levels (Fig. 3E), consistent withour findings in 52 resected patient lung SCC samples (Fig. 3Band C). Correspondingly, high expressing STXBP4 cells hadelevated DNp63 protein levels, but not mRNA levels. Addition-ally, the relative increase in PDGFRA mRNA expression is moremarked in STXBP4 low-expressing EBC-1 cells compared withSTXBP4 high-expressing RERF-LC-Sq1 cells (SupplementaryFig. S5A and S5B). The results indicate that STXBP4 regulatesPDGFRa expression in lung SCC, most likely in a DNp63-dependent manner, and also suggest that STXBP4 may serveas a new SCC biomarker.

STXBP4 depletion represses lung SCC tumor growth in vivoWe next examined the oncogenic role of STXBP4 in regulating

the expression of PDGFRa in a lung SCC cell line using a loss-of-function approach. Two independent siRNAs against STXBP4(STXBP4#1 and STXBP4#2) were transfected into the lung SCCcell line RERF-LC-Sq1, and a siRNA targeting luciferase (siLUC)was used as a control. The STXBP4 knockdown cells showed lowexpression of STXBP4, which correlated with downregulation ofboth PDGFRA mRNA (Fig. 4A) and PDGFRa protein levels (Fig.4B), consistent with our findings in resected patient lung SCCsamples.

In order to evaluate the functional relevance of STXBP4and PDGFRa expression during tumor formation, we moni-tored the colony formation of RERF-LC-Sq1 cells lentivirallytransduced with STXBP4, DNp63, or PDGFRa shRNAs. Asshown in Fig. 4C, PDGFRa, STXBP4, or DNp63 knockdownin RERF-LC-Sq1 cells led to decreased anchorage-independentcolony formation in soft agar. Subcutaneous transplantationof PDGFRA or STXBP4 knockdown clones into immunode-ficient mice resulted in suppressed tumor formation comparedwith control luciferase shRNA xenografts (Fig. 4D). Addition-ally, the knockdown effect of STXBP4 and suppression oftumorigenesis are more marked in high STXBP4-expressingRERF-LC-Sq1 cells compared with low STXBP4-expressingEBC-1 cells (Fig. 4 and Supplementary Fig. S6). These data

Table 2. Univariate and multivariate survival analysis in all patients

OS PFSUnivariate Multivariate Univariate Multivariate

Variables 5-yr rate (%) P HR (95% CI) P 5-yr rate (%) P HR (95% CI) P

Age, yr 0.853 0.504 0.965 0.452�65 49 1.32 44 1.34>65 55 (0.60–3.21) 51 (0.64–3.06)

Sex 0.579 0.690 0.215 1.90Male 55 1.25 51 1.90Female 38 (0.36–3.26) 21 (0.63–4.55)

Differentiation 0.206 0.447WD 68 54MD/PD 50 48

T factor 0.011 0.011T1 75 75T2–3 44 44

N factor 0.207 0.011N0 57 56N1–2 43 29

Disease stage 0.006 0.033 0.002 0.002I 70 2.17 67 2.94II–III 42 (1.06–4.24) 29 (1.49–5.67)

Lymphatic permeation 0.448 0.182Positive 50 44Negative 62 54

Vascular invasion 0.239 0.365Positive 49 48Negative 57 51

Pleural involvement 0.014 0.049Positive 41 41Negative 65 56

STXBP4 0.005 0.028 0.006 0.040Positive 41 2.24 37 2.02Negative 72 (1.09–5.10) 66 (1.03–4.24)

DNp63 0.362 0.458Positive 51 45Negative 61 54

p63 0.523 0.111Positive 53 51Negative 54 36






B

C

E

High Low

Positive STXBP4 Negative STXBP4

L101

L100

L106

L082

L078

L092

L075

L060

L080

L095

L050

L087

PDGFRA PDGFRB

VEGFR1 (FLT1) VEGFR2 (KDR) VEGFR3 (FLT4)

A

2.92 –2.92

Posi

tive

STXB

P4L100L101L106L078L082L092L050L060L075L080L087L095

Neg

ativ

e ST

XBP4

Z-score

Lung squamous cell carcinoma

RNA-sequencing

Positive STXBP4(n = 6)

Negative STXBP4(n = 6)

93 Genes ↓79 Genes ↑

P < 0.05FDR < 0.5

D

* FDR < 0.1

Mol

ecul

ar a

nd c

ellu

lar f

unct

ion

0 10 20 30 40

Cellular function and maintenance

Cellular movement

Cell death and survival

Cell to cell signaling and interaction

Cell morphology

(%)

(P = 1.04E–7−6.88E–3)

(P = 6.41E–7−7.49E–3)

(P = 7.94E –7−7.62E–3)

(P = 2.33E –6−7.06E–3)

(P = 7.52E –6−6.82E–3)

XAGE1BXAGE1E

DSG1TENM1

FA2HCOL4A3

DPP4NT5E

MIR100HGSLC2A3

SCG2GREM1

LGI2EPHA3

PDGFRASEMA3D

NELL2VEGFCKCNIP3

CELPCEL

RARRES1SFRP1

H19KIT

IGF2TF

−10 −5 0 5 10

log2 (fold change)

Figure 2.

Gene expression profiling of clinical samples from lung SCC patients. A, A cluster diagram of RNA-seq data from six pairs of STXBP4-positive and STXBP4-negativesamples. The color bars represent relative expression levels: red indicates higher than average expression and blue indicates lower than average expression.B, 79 significantly differentially expressed upregulated genes and 93 downregulated genes (P < 0.05 and FDR < 0.5) were identified. C, Functional analysis ofdifferentially expressed genes was performed by Ingenuity Pathway Analysis. D, The genes are listed in descending order of normalized expression. � , FDR< 0.1. E, A cluster diagram of PDGFR and VEGFR expression from RNA-seq analysis. The color bars represent relative expression levels: red indicateshigher than average expression and blue indicates lower than average expression.

Otaka et al.





suggest that downregulation of STXBP4 decreases PDGFRaexpression and suppresses tumor formation.

Overall, our results indicate that STXBP4 has oncogenic activityboth in vitro and in vivo and further suggest that STXBP4 could be acritical driver of tumor propagation through regulating thePDGFRa pathway.

DiscussionWedemonstrated that STXBP4 expression in clinical specimens

was closely associated with T factor (P < 0.001), disease stage (P¼0.030), and pleural involvement (P ¼ 0.028). Furthermore,

univariate and multivariate analysis indicated that STXBP4expression was an independent prognostic factor for OS and PFS.While p63 is essential for normal epidermal stratification and theproliferative potential of epithelial stem cells, DNp63 is thoughtto maintain the proliferative potential of basal regenerative cells,including stem cells, in skin, thymus, breast, prostate, and urothe-lial stratified epithelium (20, 32–36). STXBP4 can physicallyinteract with DNp63 and is indispensable for stabilizing DNp63,which is consistent with a putative diagnostic role for STXBP4 inlung SCC.

Polymorphisms of STXBP4/COX11 (rs6504950; AA/AG-geno-type) were associated with a significantly decreased risk of

B C

A

STXB

P4

Positive STXBP4 (n = 34) Negative STXBP4 (n = 18)

L004

L049

L047

L024

L051

L002

L090

L015

L076

L019

L091

L069

L059

L102

L061

L031

L081

L084

L077

L074

L057

L083

L097

L066

L098

L103

L100

L101

L106

L088

L082

L092

L099

L078

L033

L067

L062

L060

L050

L017

L075

L094

L048

L104

L087

L085

L016

L079

L080

L095

L107

L096

(n = 52)

0

1

2

3

4

5

PDGFRAPDGFRBVEGFR1VEGFR2VEGFR3

LowHigh

Rel

ativ

e m

RN

A le

vels

VEGFR1 (FLT1) VEGFR2 (KDR) VEGFR3 (FLT4)

PDGFRA PDGFRB

*P < 0.05

Nega. Posi. Nega. Posi.

Nega. Posi. Nega. Posi. Nega. Posi.

D E

0

2

4

6

Moc

k

STXB

P4

DNp63

0

1

Moc

k

STXB

P4

DNp6

3a

DNp6

3a

DNp6

3a

DNp6

3a DNp63a

2

3

4

5PDGFRA

Rel

ativ

e m

RN

A le

vels

STXBP4

Moc

k

STXB

P4

p-PDGFRa

b-Actin

PDGFRa

p-p38MAPK

p38MAPK

R = 0.610 (95%CI, 0.397−0.761)*P < 0.01

STXBP4

PDG

FRA

(n = 52)

STXBP4

0

2

4

6

8

Moc

k

STXB

P4

Figure 3.

PDGFRA expression is significantlyupregulated in STXBP4-positivesamples from lung SCC patients.A, A cluster diagram of relativeexpression of PDGFRs and VEGFRsby real-time RT-PCR analysis. Tumorspecimens were collected from lungSCC patients with surgical resection.A total of 52 high RNA integritynumber (RIN > 2.0) STXBP4-positivesamples (n ¼ 34) and STXBP4-negative samples (n ¼ 18) were usedfor transcriptome profiling by real-time RT-PCR. B, PDGFRAmRNA wassignificantly upregulated in STXBP4-positive lung SCC samples. RelativemRNA levels of PDGFRs and VEGFRsof the STXBP4-positive samples(n ¼ 34) and STXBP4-negativesamples (n ¼ 18); � , P < 0.05. C,Scatter plot of relative mRNAexpression levels of PDGFRA andSTXBP4. D, STXBP4 or DNp63ainduces PDGFRA expression in a lungSCC cell line, EBC-1. The cells wereretrovirally transduced with emptyvector control (Mock), DNp63a orSTXBP4. The mRNA levels ofPDGFRA, STXBP4, or DNp63 weredetermined by real-time RT-PCR. E,EBC-1 cells transduced as in D weresubjected to immunoblotting usingantibodies for phospho-PDGFRa(p-PDGFRa), PDGFRa, phospho-p38MAPK (p-p38MAPK), p38MAPK,STXBP4, DNp63 or b-Actin.






carcinogenesis in a meta-analysis of breast cancer patients (37).Although functional assessments of these polymorphisms werenot undertaken, and the extent of loss of STXBP4 function intumors was not studied, the data suggested that STXBP4 couldplay a role in carcinogenesis and tumor progression in breastcancer patients. Our study provides diagnostic role of STXBP4alongside DNp63 in lung SCC.

The pathologic function of STXBP4 in human cancers remainsunclear. However, STXBP4 can physically interact with p63 and isindispensable for stabilizing DNp63 even in normal conditions(16). Consistent with STXBP4 localization in both the nucleusand cytoplasm, it has been suggested that nuclear STXBP4 hasp63-mediated functions, and that cytoplasmic STXBP4 couldfacilitate other functions in a p63-independent manner (16). Infact, our data indicated that STXBP4 induction partially increasedtumor growth even in the absence of elevated DNp63 (Supple-mentary Fig. S7A), and that PDGFRA induction also partiallyincreased tumor growth even in STXBP4 knockdown cells (Sup-plementary Fig. S7B). Thus, STXBP4 may contribute to the sus-ceptibility and severity of cancer in a p63-dependent and -inde-

pendent manner. Indeed, amplification and overexpression ofp63 has frequently been observed in a variety of SCCs, includinglung cancers and head and neck cancers (8, 38). However, p63expression is decreased during progression to invasion andmetas-tasis of lung, breast, and bladder cancers, and loss of p63 expres-sion is associatedwithworse prognosis in some cases (35, 39, 40).It could be the balance between the TA isotype (tumor suppres-sive) and DN isotype (oncogenic), as well as the tissue context,which is critical for proliferation and differentiation in bothepithelial stem cells and cancer stem cells.

Global transcriptome profiling using next-generation sequenc-ing technologies has become more common for comprehensivegene expression analysis to explore novel regulators and targetgenes in different types of cancers. In this report, genome-widetranscriptome analysis identified mediators of STXBP4 activity,including PDGFRa, which contributes to cell growth and metas-tasis in a p63-dependentmanner. PDGF family proteins consist ofseveral disulfide-bonded, dimeric isoforms (PDGF AA, PDGF AB,PDGFBB, PDGFCC, andPDGFDD) that bind in a specific patternto two related receptor tyrosine kinases, PDGFRa and PDGFRb

BA

C D

PDGFRa

DNp63

b-Actin

STXBP4

siLU

C

STXB

P4 #

1

DNp6

3

STXB

P4 #

2

PDG

FRa

siRNA:

0

0.5

1.0

shLU

C

STXB

P4

DNp6

3

PDG

FRa

shRNA:

**

*

Col

ony

form

atio

n

shLUC STXBP4 DNp63 PDGFRa

shRNA

Rel

ativ

e m

RN

A e

xpre

ssio

n

0

0.5

1.0

siRNA:

1.5PDGFRA

siLU

C

STXB

P4#1

DNp6

3

STXB

P4#2

PDG

FRa

** * *

Tum

or v

olum

e (m

m3 )

0

200

400

600

800

1,000

shLU

C

STXB

P4

DNp6

3

PDG

FRa

shRNA:

**

*

shRNA

shLUC STXBP4 DNp63 PDGFRD

Figure 4.

STXBP4-depletion inhibits lung SCCtumorigenesis and modulates PDGFsignaling in vivo. A, The lung SCC cells,RERF-LC-Sq1, were treated withsiRNAs for Luciferase (siLUC) as acontrol, STXBP4#1, STXBP4#2, DNp63,or PDGFRa. Total RNAs werequantified by real-time RT-PCRanalysis, and the induction levels ofPDGFRA were determined by therelative Ct method. B, RERF-LC-Sq1cells depleted as inAwere subjected toimmunoblotting using anti-PDGFRa,STXBP4, DNp63, or b-Actin antibodies.C, The growth of RERF-LC-Sq1 cellsafter shRNA mediated PDGFRa,STXBP4, or DNp63 knockdown wasmonitored by soft-agar colonyformation assays. Standard deviations(SD) are plotted. � , P < 0.05. D,Representative images of xenograftssubcutaneously transplanted withlentivirally shRNA transducedLuciferase as a control (shLUC),Stxbp4, DNp63, or PDGFRaknockdown RERF-LC-Sq1 cells(n¼6 for each knockdown). The resultsof six independent injections ofknockdown cells are shown. Twentydays after implantation, the length (L)and width (W) of the tumor mass weremeasured, and the tumor volume (TV)was calculated using the equation:TV ¼ (L � W2)/2. � , P < 0.05.

Otaka et al.





(41). PDGFRa homodimers bind to all PDGF isoforms exceptthose containing PDGF D (42). A number of different signalingpathways, including mTOR (Fig. 1D) and MAPK (Fig. 3E), areinitiated by activated PDGF receptors, and stimulate cell growth,actin reorganization, migration, and differentiation (43, 44).

PDGF receptors are expressed at low levels in normal lungepithelial cells; however, increased PDGFRa expression has beenreported in lung cancer. PDGFRb expression is observedmainly instromal cells, but also in the sarcomatoid type of NSCLC (45).Based on recent evidence, inhibition of the p53/NF-Y complex bymutant gain-of-function p53 enhances PDGFRb expression andpromotes metastasis in a subset of pancreatic cancers (31). Inaddition, the interaction of mutant p53 with p63 regulates theexpression of p63 target genes to enhance invasion andmetastasis(46). Hence, the oncogenic activity of mutant p53 is a conse-quence of the physical association between mutant p53 and thep53 family members p63 and p73.

Treatment strategies for lung cancer are based on theassumption that an individual patient's cancer is purely ofone subtype. Because many cancers are heterogeneous andrelatively resistant to chemotherapy or radiation, there isstrong interest in molecular-targeted therapies based on tumorbiology. In particular, targeted agents that inhibit the epider-mal growth factor receptor (EGFR) or anaplastic lymphomakinase (ALK) are approved for the treatment of NSCLC har-boring genetic alterations in the genes encoding these proteins(47). EGFR inhibitors, such as erlotinib and gefitinib, are onlyeffective against NSCLCs with EGFR mutations, which occuralmost exclusively in lung AC. Similarly, the recently identifiedEML4–ALK rearrangement, which predicts susceptibility to thetargeted agent crizotinib, also occurs only in lung AC. Unfor-tunately, therapeutic advances in the treatment of lung SCChave lagged behind those for AC (48). Therefore, the capacityto distinguish between lung AC and SCC is particularly impor-tant for the effective use of novel targeted therapies to treatpatients with these NSCLC subtypes.

Inhibition of the PDGFRa signaling pathway by treatmentwitha neutralizing PDGFRa antibody, MEDI-575, had minimal effecton tumor cell proliferation in preclinical models of NSCLC (49).Lung SCC histology also identified patients at a higher risk ofbleeding during treatment with bevacizumab, amonoclonal anti-VEGFantibody (50). Thus,more studies are required todeterminewhether specific inhibition of PDGF receptors, without inhibitionof VEGF receptors, is of any benefit for lung cancer patients. Theseissues highlight the growing importance of accurate identification

of NSCLC subtypes for assigning patients to appropriate histol-ogy-based therapies and the triage of tissue for appropriatemolecular studies.

Disclosure of Potential Conflicts of InterestM.Nishiyama receives commercial research support fromYakultHonshaCo.

Ltd. No potential conflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: Y. Otaka, S. Rokudai, J. Tamura, C. Prives,M. NishiyamaDevelopment of methodology: Y. Otaka, S. Rokudai, K. Kaira, Y. OhtakiAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): Y. Otaka, S. Rokudai, M. Fujieda, I. Horikoshi,K. Shimizu, M. NishiyamaAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): Y. Otaka, S. Rokudai, K. Kaira, R. Iwakawa-Kawabata,S. YoshiyamaWriting, review, and/or revision of the manuscript: Y. Otaka, S. Rokudai,K. Kaira, R. Iwakawa-Kawabata, Y. Ohtaki, K. Shimizu, M. NishiyamaAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): Y. Otaka, S. Rokudai, R. Iwakawa-Kawabata,S. Yoshiyama, K. Shimizu, C. PrivesStudy supervision: S. Rokudai, T. Yokobori, T. Oyama, J. Tamura, C. Prives,M. Nishiyama

AcknowledgmentsWe thankDr. Arito Yamane for instrument settings of the Ion Proton system,

andN. Gombodorj, S. Umezawa, Y. Suto,M. Nakamura,M. Ito, A. Ichihara, andE. Horigome for expert technical assistances. We also thank Drs. VincenzoCastronovo, Andrei Turtoi, and Akeila Bellahcene in the University of Liege forvaluable discussions.

Grant SupportThis work was supported, in part, by the Promotion Plan for the Platform of

Human Resource Development for Cancer, New Paradigms – EstablishingCenter for Fostering Medical Researchers of the Future Programs by Ministryof Education, Culture, Sports, Science and Technology of Japan, GunmaUniversity Initiative for Advanced Research (GIAR), Promotion Plan for thePlatform of Human Resource Development for Cancer, Technology of Japan,and Yasuda Memorial Medical Foundation. This work was also supported byNIH grant CA87497.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

Received July 22, 2016; revised November 20, 2016; accepted December 30,2016; published OnlineFirst January 13, 2017.

References1. Molina JR, Yang P, Cassivi SD, Schild SE, Adjei AA. Non-small cell lung

cancer: epidemiology, risk factors, treatment, and survivorship. Mayo ClinProc 2008;83:584–94.

2. Recondo GJr, Recondo GSr, Galanternik F, Greco M, de la Vega M, CantonED, et al. Immunotherapy for non-small cell lung cancer – finally a hint ofhope. Rev Recent Clin Trials 2016;11:87–92.

3. Mukhopadhyay S, Katzenstein AL. Subclassification of non-small cell lungcarcinomas lacking morphologic differentiation on biopsy specimens:utility of an immunohistochemical panel containing TTF-1, napsin A,p63, and CK5/6. Am J Surg Pathol 2011;35:15–25.

4. Massion PP, Taflan PM, Jamshedur Rahman SM, Yildiz P, Shyr Y, EdgertonME, et al. Significance of p63 amplification and overexpression in lungcancer development and prognosis. Cancer Res 2003;63:7113–21.

5. Sniezek JC, Matheny KE, Westfall MD, Pietenpol JA. Dominant negativep63 isoform expression in head and neck squamous cell carcinoma.Laryngoscope 2004;114:2063–72.

6. Candi E, Dinsdale D, Rufini A, Salomoni P, Knight RA, Mueller M, et al.TAp63 and DeltaNp63 in cancer and epidermal development. Cell Cycle2007;6:274–85.

7. Mills AA. p63: oncogene or tumor suppressor? Curr Opin Genet Dev 2006;16:38–44.

8. DiComoCJ,UristMJ, Babayan I,DrobnjakM,HedvatCV, Teruya-FeldsteinJ, et al. p63 expression profiles in human normal and tumor tissues. ClinCancer Res 2002;8:494–501.

9. Nonaka D. A study of DeltaNp63 expression in lung non-small cellcarcinomas. Am J Surg Pathol 2012;36:895–9.






10. Bishop JA, Teruya-Feldstein J, Westra WH, Pelosi G, Travis WD, RekhtmanN. p40 (DeltaNp63) is superior to p63 for the diagnosis of pulmonarysquamous cell carcinoma. Mod Pathol 2012;25:405–15.

11. MelinoG, Lu X,GascoM, Crook T, Knight RA. Functional regulation of p73and p63: development and cancer. Trends Biochem Sci 2003;28:663–70.

12. Yang A, Kaghad M, Wang Y, Gillett E, Fleming MD, Dotsch V, et al. p63,a p53 homolog at 3q27-29, encodes multiple products with transacti-vating, death-inducing, and dominant-negative activities. Mol Cell1998;2:305–16.

13. Schmale H, Bamberger C. A novel protein with strong homology to thetumor suppressor p53. Oncogene 1997;15:1363–7.

14. McKeon F. p63 and the epithelial stem cell: more than status quo? GenesDev 2004;18:465–9.

15. Duijf PH, Vanmolkot KR, Propping P, Friedl W, Krieger E, McKeon F,et al. Gain-of-function mutation in ADULT syndrome reveals thepresence of a second transactivation domain in p63. Hum Mol Genet2002;11:799–804.

16. Li Y, Peart MJ, Prives C. Stxbp4 regulates DeltaNp63 stability by suppres-sion of RACK1-dependent degradation. Mol Cell Biol 2009;29:3953–63.

17. Fomenkov A, Zangen R, Huang YP, Osada M, Guo Z, Fomenkov T, et al.RACK1 and stratifin target DeltaNp63alpha for a proteasome degradationin head and neck squamous cell carcinoma cells upon DNA damage. CellCycle 2004;3:1285–95.

18. Min J,Okada S, KanzakiM,Elmendorf JS, CokerKJ, Ceresa BP, et al. Synip: anovel insulin-regulated syntaxin 4-binding protein mediating GLUT4translocation in adipocytes. Mol Cell 1999;3:751–60.

19. Saito T, Okada S, Yamada E, Ohshima K, Shimizu H, Shimomura K,et al. Syntaxin 4 and Synip (syntaxin 4 interacting protein) regulateinsulin secretion in the pancreatic beta HC-9 cell. J Biol Chem 2003;278:36718–25.

20. Westfall MD, Mays DJ, Sniezek JC, Pietenpol JA. The Delta Np63 alphaphosphoprotein binds thep21 and14-3-3 sigmapromoters in vivo andhastranscriptional repressor activity that is reduced by Hay-Wells syndrome-derived mutations. Mol Cell Biol 2003;23:2264–76.

21. Shimizu K, Kaira K, Tomizawa Y, SunagaN, KawashimaO,Oriuchi N, et al.ASC amino-acid transporter 2 (ASCT2) as a novel prognostic marker innon–small cell lung cancer. Br J Cancer 2014;110:2030–9.

22. Kaira K, Sunose Y, Arakawa K, Sunaga N, Shimizu K, Tominaga H, et al.Clinicopathological significance of ASC amino acid transporter-2expression in pancreatic ductal carcinoma. Histopathology 2015;66:234–43.

23. Rokudai S, Aikawa Y, Tagata Y, Tsuchida N, Taya Y, Kitabayashi I. Mono-cytic leukemia zinc finger (MOZ) interacts with p53 to induce p21expression and cell-cycle arrest. J Biol Chem 2009;284:237–44.

24. Rokudai S, Laptenko O, Arnal SM, Taya Y, Kitabayashi I, Prives C. MOZincreases p53 acetylation and premature senescence through its complexformation with PML. Proc Natl Acad Sci U S A 2013;110:3895–900.

25. Trapnell C, Roberts A, Goff L, Pertea G, KimD, Kelley DR, et al. Differentialgene and transcript expression analysis of RNA-seq experiments withTopHat and Cufflinks. Nat Protoc 2012;7:562–78.

26. Anders S, McCarthy DJ, Chen Y, Okoniewski M, Smyth GK, Huber W, et al.Count-based differential expression analysis of RNA sequencing data usingR and Bioconductor. Nat Protoc 2013;8:1765–86.

27. Zoncu R, Efeyan A, Sabatini DM.mTOR: from growth signal integration tocancer, diabetes and ageing. Nat Rev Mol Cell Biol 2011;12:21–35.

28. HeldinCH. Targeting thePDGF signaling pathway in tumor treatment.CellCommun Signal 2013;11:97.

29. Dai Y. Platelet-derived growth factor receptor tyrosine kinase inhibitors:a review of the recent patent literature. Expert Opin Ther Pat 2010;20:885–97.

30. CaoR, BjorndahlMA, Religa P,Clasper S,Garvin S,GalterD, et al. PDGF-BBinduces intratumoral lymphangiogenesis and promotes lymphatic metas-tasis. Cancer Cell 2004;6:333–45.

31. Weissmueller S, Manchado E, Saborowski M, Morris JPt, Wagenblast E,Davis CA, et al. Mutant p53 drives pancreatic cancer metastasis throughcell-autonomous PDGF receptor beta signaling. Cell 2014;157:382–94.

32. Mills AA, Zheng B, Wang XJ, Vogel H, Roop DR, Bradley A. p63 is a p53homologue required for limb and epidermal morphogenesis. Nature1999;398:708–13.

33. Yang A, Schweitzer R, SunD, KaghadM,Walker N, Bronson RT, et al. p63 isessential for regenerative proliferation in limb, craniofacial and epithelialdevelopment. Nature 1999;398:714–8.

34. Tonon G, Wong KK, Maulik G, Brennan C, Feng B, Zhang Y, et al. High-resolution genomic profiles of human lung cancer. ProcNatl Acad Sci U S A2005;102:9625–30.

35. Barbieri CE, Tang LJ, BrownKA, Pietenpol JA. Loss of p63 leads to increasedcell migration and up-regulation of genes involved in invasion andmetastasis. Cancer Res 2006;66:7589–97.

36. Higashikawa K, Yoneda S, Tobiume K, Taki M, Shigeishi H, Kamata N.Snail-induced down-regulation of DeltaNp63alpha acquires invasive phe-notype of human squamous cell carcinoma. Cancer Res 2007;67:9207–13.

37. Tang L, Xu J,Wei F,Wang L,NieWW,Chen LB, et al. Association of STXBP4/COX11 rs6504950 (G>A) polymorphism with breast cancer risk: evidencefrom 17,960 cases and 22,713 controls. Arch Med Res 2012;43:383–8.

38. Yamaguchi K, Wu L, Caballero OL, Hibi K, Trink B, Resto V, et al. Frequentgain of the p40/p51/p63 gene locus in primary head and neck squamouscell carcinoma. Int J Cancer 2000;86:684–9.

39. Karni-Schmidt O, Castillo-Martin M, Shen TH, Gladoun N, Domingo-Domenech J, Sanchez-CarbayoM, et al. Distinct expression profiles of p63variants during urothelial development and bladder cancer progression.Am J Pathol 2011;178:1350–60.

40. Deyoung MP, Ellisen LW. p63 and p73 in human cancer: defining thenetwork. Oncogene 2007;26:5169–83.

41. Deuel TF, Silverman NJ, Kawahara RS. Platelet-derived growth factor: amultifunctional regulator of normal and abnormal cell growth. BioFactors1988;1:213–7.

42. Bergsten E,UutelaM, Li X, Pietras K,OstmanA,HeldinCH, et al. PDGF-D isa specific, protease-activated ligand for the PDGF beta-receptor. Nat CellBiol 2001;3:512–6.

43. Ostman A, Heldin CH. Involvement of platelet-derived growth factorin disease: development of specific antagonists. Adv Cancer Res 2001;80:1–38.

44. Andrae J, Gallini R, Betsholtz C. Role of platelet-derived growth factors inphysiology and medicine. Genes Dev 2008;22:1276–312.

45. Donnem T, Al-Saad S, Al-Shibli K, Andersen S, Busund LT, Bremnes RM.Prognostic impact of platelet-derived growth factors in non-small cell lungcancer tumor and stromal cells. J Thorac Oncol 2008;3:963–70.

46. AdornoM,CordenonsiM,MontagnerM,Dupont S,WongC,HannB, et al.A Mutant-p53/Smad complex opposes p63 to empower TGFbeta-inducedmetastasis. Cell 2009;137:87–98.

47. Solomon BJ, Mok T, Kim DW, Wu YL, Nakagawa K, Mekhail T, et al. First-line crizotinib versus chemotherapy in ALK-positive lung cancer. N Engl JMed 2014;371:2167–77.

48. Socinski MA, Stinchcombe TE, Moore DT, Gettinger SN, Decker RH, PettyWJ, et al. Incorporating bevacizumab and erlotinib in the combined-modality treatment of stage III non-small-cell lung cancer: results of aphase I/II trial. J Clin Oncol 2012;30:3953–9.

49. Laing N, McDermott B, Wen S, Yang D, Lawson D, Collins M, et al.Inhibition of platelet-derived growth factor receptor alpha by MEDI-575reduces tumor growth and stromal fibroblast content in a model of non-small cell lung cancer. Mol Pharmacol 2013;83:1247–56.

50. Johnson DH, Fehrenbacher L, Novotny WF, Herbst RS, Nemunaitis JJ,Jablons DM, et al. Randomized phase II trial comparing bevacizumab pluscarboplatin and paclitaxel with carboplatin and paclitaxel alone in previ-ously untreated locally advancedormetastatic non-small-cell lung cancer. JClin Oncol 2004;22:2184–91.


Otaka et al.




2017;23:3442-3452. Published OnlineFirst January 13, 2017.Clin Cancer Res Yukihiro Otaka, Susumu Rokudai, Kyoichi Kaira, et al. Cell CarcinomaPrognosis through PDGF Receptor Signaling in Lung Squamous STXBP4 Drives Tumor Growth and Is Associated with Poor

Updated version

10.1158/1078-0432.CCR-16-1815doi:

Access the most recent version of this article at:

Material

Supplementary

http://clincancerres.aacrjournals.org/content/suppl/2017/01/12/1078-0432.CCR-16-1815.DC1

Access the most recent supplemental material at:

Cited articles

http://clincancerres.aacrjournals.org/content/23/13/3442.full#ref-list-1

This article cites 50 articles, 15 of which you can access for free at:

Citing articles

http://clincancerres.aacrjournals.org/content/23/13/3442.full#related-urls

This article has been cited by 4 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/23/13/3442To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/lookup/doi/10.1158/1078-0432.CCR-16-1815

http://clincancerres.aacrjournals.org/content/suppl/2017/01/12/1078-0432.CCR-16-1815.DC1

http://clincancerres.aacrjournals.org/content/23/13/3442.full#ref-list-1

http://clincancerres.aacrjournals.org/content/23/13/3442.full#related-urls

http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://clincancerres.aacrjournals.org/content/23/13/3442


stxbp4 drives tumor growth and is associated with poor … · biology of human tumors stxbp4 drives...

Documents